Re: st: Ordered Logit - Should I use it or not

[Sunil Kumar <stuff.sunil@googlemail.com>]

Augusto,

I see no reason why you should not simply treat your dependent
variable--willingness to pay--as a continuous variable, censored at 0.
That means you can analyze it using a tobit model, or, if you believe
that the process determining the binary decision to be willing to pay
anything (yes or no) is different from the process that determines the
amount, conditional on that amount being positive, then you could
explore a double hurdle type model (see, for instance, Moffatt, Peter
G. 2005. Hurdle models of loan default. Journal of the Operational
Research Society 56 (9): 1063-1071.).

Best,
Sunil

On Sat, Jul 14, 2012 at 3:41 PM, Augusto Pallavicini
<agupalla@hotmail.com> wrote:
> Hi all,
> I wanted to know whether I should use ordered logit. I am working on a
> contingent valuation dissertation.
> My dependent variable is willingness to pay and contains the values 0; 2.5;
> 5; 7.5; 10; 15.
> I transformed these values into 6 categories 1 2 3 4 5 6
> The original question from which I obtained the data was Would you willing
> to pay 2.5 pounds for...? If yes, would you willing to pay 5...? and so on
> If I cannot use ordered logit, what do you recommend me to use?
> Thanks a lot
> Augusto
> *
> *   For searches and help try:
> *   http://www.stata.com/help.cgi?search
> *   http://www.stata.com/support/statalist/faq
> *   http://www.ats.ucla.edu/stat/stata/
*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/